metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: context
dtype: string
splits:
- name: train
num_bytes: 15623876.121307762
num_examples: 70719
- name: test
num_bytes: 1736059.8786922381
num_examples: 7858
download_size: 8588232
dataset_size: 17359936
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
Dataset diviso in train e test (dataset originale -> "b-mc2/sql-create-context")